If the resources available in the cluster of Avaya Breeze® platform servers is not sufficient for the configured deployment (data grid size plus optional features selected), the deployment fails. For attribute values for each certified deployment, see Avaya Context Store Snap-in Release Notes.
If deployment issues occur, check your configuration settings are correct, check the /var/log/Avaya/services/ContextStoreManager and /var/log/Avaya/dcm/gs-esm-<X> logfiles for errors related to deployment.
The following errors indicate there has been a data-grid deployment failure:
-
Instead of the ContextStoreManager status page located at : http://<IP_ADDRESS>/services/ContextStoreManager/, the following error is displayed: Error 503: Internal error Could not read statistics object, check CSSpace settings. EDM PU deployment failed, check EDM settings, or try accessing the homepage via load balancer IP or other node GEO PU deployment failed, review GEO settings, or try accessing the homepage via load balancer IP or other node.
-
A 404 not found response is returned for non-read request. For example, Error 404: javax.servlet.UnavailableException: Space ContextStoreManagerSpace could not be found in 60000 milliseconds to create a listener.
-
The following error is displayed: Number of retries (100) to create proxy to grid has been exhausted, machine restart required.
Note:
It is imperative that nodes are shut down at the same time to ensure that the data-grid is fully undeployed before the next deployment attempt takes place. The last node standing will keep the data-grid alive. So when nodes have already started back up before that last-man-standing shuts down, the data grid does not get redeployed. This necessity is stated in the documentation, this is what we mean by “simultaneous restart”.
-
After a correct fresh start of the cluster, you will see these lines in the ContextStoreManager log file (on whichever node has taken charge of the deployment. Whoever got there first)
-
grid.EsmGrid INFO - [M eployElasticPU][T:null]. ContextStoreSpace not found, deploying...
-
grid.EsmGrid INFO - [M:deployStateless][T:null]. EDM-Mirror-Service is deployed with status BROKEN.
-
Followed quickly by manager.AbstractPUManager INFO - [M:checkDeploymentStatus][T:null]. [EDM-Mirror-Service] is in status [INTACT] a few seconds later
-
If these deploying messages do not appear on any ContextStoreManager logs since restart and instead you get already deployed messages, you have not restarted the cluster correctly.
-
grid.EsmGrid INFO - [M eployElasticSpace][T:null]. space with name ContextStoreManagerSpace is already deployed.
-
grid.EsmGrid ERROR - [M:deployElasticPU][T:null]. space ContextStoreSpace is already deployed.
